An online recognition method for the warping degree of cardboard at the outlet of a cardboard production line

Abstract

The present invention is applicable to the field of cardboard production quality control, and discloses an on-line recognition method for the warping degree of cardboard at the outlet of a cardboard production line, including collecting a cross-sectional image of the cardboard; calculating the warping convexity of the cardboard according to the cross-sectional image; judging the quality of the cardboard according to the warping convexity. If the absolute value of the warping convexity is not greater than a preset threshold, it is determined that the quality of the cardboard is qualified. If the absolute value of the warping convexity is greater than the preset threshold, the type and level of the cardboard warping are determined and output according to the value of the warping convexity and the preset range values of the warping convexity corresponding to different warping types and levels. This method realizes the on-line recognition of the cardboard warping degree through the means of quantifying the warping degree, solves the problem that the existing cardboard at the outlet of the cardboard production line relies on manual recognition of the warping degree, and lays a foundation for realizing the process control of the cardboard quality at the export end of the cardboard.

Description

Technical Field

[0001] The present invention relates to the field of cardboard production quality control, and in particular to an online recognition method for the warping degree of cardboard at the outlet of a corrugated cardboard production line. Background Art

[0002] Corrugated paper packaging is widely used in downstream consumer industries such as home appliances, electronic products, IT, food and beverage, books, daily chemicals, textiles, etc. and logistics and express companies. The production of corrugated cardboard consists of key links such as the corrugation of the base paper, the formation of the cardboard, and the cutting and stacking. The quality control and efficiency guarantee not only involve the data analysis and intelligent control of the equipment used in the key links, but also involve the effective process control problem of the corrugated cardboard production line.

[0003] The warping degree of the cardboard at the outlet is an important indicator reflecting the product quality of the corrugated cardboard production line. The product quality of the corrugated cardboard production line is mainly related to factors such as the base paper material, moisture content, and temperature control during the production process. The process control of the corrugated cardboard production line is mainly to monitor the warping degree of the cardboard at the outlet of the production line in real time and online, and achieve effective feedback control of the cardboard quality. Among them, the online detection of the warping degree of the cardboard at the outlet of the production line and the recognition of the warping level are the keys to realizing the process control of the production line. At present, the quality inspection and control of the export products of the corrugated cardboard production lines that have been put into production in China mainly adopt the method of on-site engineers visually judging and recognizing, and manually inputting parameters to adjust and control the product quality. However, the current production speed of the corrugated cardboard production line is generally several hundred meters per minute. It is difficult to rely on manual labor to achieve the online detection of the warping degree of the cardboard at the outlet and the recognition of the warping level; moreover, the rapid development of industries such as online shopping and express delivery has brought customized and personalized market demands, resulting in the requirement of multiple batches and small quantities for corrugated cardboard production. The number of orders is large and the frequency of material changes during the production process is very high, resulting in a high labor cost in the corrugated cardboard industry, which affects the automation and intelligent level of the corrugated cardboard manufacturing industry. Summary of the Invention

[0004] The purpose of the present invention is to provide an online recognition method for the warping degree of cardboard at the outlet of a cardboard production line, which aims to solve the technical problem that the existing cardboard production line depends on manual recognition of the warping degree of the cardboard at the outlet.

[0005] To achieve the above object, the solution provided by the present invention is:

[0006] An online recognition method for the warping degree of cardboard at the outlet of a cardboard production line, comprising:

[0007] Collecting a cross-sectional image of the cardboard;

[0008] Calculating the warping convexity of the cardboard according to the cross-sectional image;

[0009] Judge the quality of the cardboard according to the warping convexity. If the absolute value of the warping convexity is not greater than the preset threshold, it is determined that the quality of the cardboard is qualified. If the absolute value of the warping convexity is greater than the preset threshold, the type and level of the cardboard warping are determined and output according to the value of the warping convexity and the preset range values of the warping convexity corresponding to different warping types and levels.

[0010] Preferably, the cross-sectional image is collected by a camera, and the camera is arranged at the outlet end of the cardboard production line.

[0011] Preferably, the calculation of the warping convexity of the cardboard includes:

[0012] Extract the cross-sectional contour of the cardboard according to the cross-sectional image;

[0013] Perform data fitting on the extracted cross-sectional contour to obtain a quadratic function relationship of the cross-sectional contour transverse pixel coordinate - cross-sectional contour longitudinal pixel coordinate. The quadratic function relationship is:

[0014] f(x) = ax 2 + bx + c (1)

[0015] where x represents the transverse pixel coordinate of the cross-sectional contour, f(x) represents the longitudinal pixel coordinate of the cross-sectional contour at the corresponding x position, and a, b, c represent constants;

[0016] Calculate the warping convexity of the cardboard based on the quadratic function relationship.

[0017] Preferably, the calculation of the warping convexity of the cardboard further includes performing image processing on the cross-sectional image before extracting the cross-sectional contour of the cardboard. The image processing includes performing image grayscale processing and region of interest extraction on the sample.

[0018] Preferably, the calculation of the warping convexity of the cardboard further includes performing image closing operation processing on the extracted cross-sectional contour before performing data fitting according to the extracted cross-sectional contour.

[0019] Preferably, the warping convexity of the cardboard is calculated using formula (2), and formula (2) is as follows:

[0020]

[0021] where δ represents the warping convexity of the cardboard, x a and x b respectively represent the pixel coordinates of the two endpoints of the transverse pixel coordinates of the quadratic function, x 1 , x 2 , x 3The pixel coordinates of the three quarter points of the horizontal pixel coordinates of the quadratic function are respectively represented.

[0022] Preferably, the method for obtaining the preset range values of the warping convexity corresponding to different warping types and warping levels is as follows:

[0023] Collect cross-sectional images of cardboard with different warping directions and degrees as samples;

[0024] Identify the warping direction and degree of the samples and classify the samples according to different warping types and warping levels to obtain classification results;

[0025] Establish sub-image libraries corresponding to different warping types and warping levels according to the classification results;

[0026] Calculate the warping convexity of each sample in each sub-image library one by one, and determine the warping convexity range values of the corresponding warping types and warping levels of each sub-image library according to the minimum and maximum values of the warping convexity of the samples in each sub-image library.

[0027] Preferably, the calculation of the warping convexity of each sample includes:

[0028] Extract the cross-sectional contour of the cardboard;

[0029] Perform data fitting according to the extracted cross-sectional contour to obtain a quadratic function relationship between the horizontal pixel coordinates of the cross-sectional contour - the vertical pixel coordinates of the cross-sectional contour. The quadratic function relationship is:

[0030] f(x') = a'x' 2 + b'x' + c' (3)

[0031] Where x' represents the horizontal pixel coordinates of the cross-sectional contour, f(x') represents the vertical pixel coordinates of the cross-sectional contour at the corresponding x' position, and a', b', c' represent constants;

[0032] Calculate the warping convexity of the sample based on the quadratic function relationship.

[0033] Preferably, the warping convexity of the sample is calculated using the formula (4). The formula (4) is as follows:

[0034]

[0035] Where δ' represents the warping convexity of the sample, x' a and x' b respectively represent the pixel coordinates at the two endpoints of the horizontal pixel coordinates of the quadratic function, x' 1 , x' 2 , x' 3They respectively represent the pixel coordinates of the three quartering points of the horizontal pixel coordinates of the quadratic function of one variable at their locations.

[0036] Preferably, the warping types include upward warping and downward warping. The levels of upward warping include at least three types, and the levels of downward warping include at least three types.

[0037] The online recognition method for the warping degree of the cardboard at the outlet of the cardboard production line provided by the present invention uses machine vision technology with the characteristics of non-contact, low cost, and high accuracy to collect the cross-sectional image of the cardboard at the outlet for calculation to obtain the warping convexity of the cardboard, and combines the preset threshold and the preset range values of the warping convexity corresponding to different warping types and warping levels to achieve the online recognition of the warping degree of the cardboard at the outlet of the cardboard production line. That is, the online recognition method for the warping degree of the cardboard at the outlet of the cardboard production line of the present invention realizes the online recognition of the warping degree of the cardboard through the means of quantifying the warping degree, solves the problem that the existing cardboard at the outlet of the cardboard production line relies on manual recognition of the warping degree, lays a foundation for realizing the process control of the quality of the cardboard at the export end of the cardboard, can effectively improve the production efficiency, and improve the stability of the cardboard production line. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on the structures shown in these drawings.

[0039] Figure 1 is a schematic flow chart of the online recognition method for the warping degree of the cardboard at the outlet of the cardboard production line provided by the embodiment of the present invention;

[0040] Figure 2 is Figure 1 a schematic flow chart of the process of calculating the warping convexity of the cardboard according to the cross-sectional image in the method;

[0041] Figure 3 is a schematic flow chart of the method for obtaining the preset range values of the warping convexity corresponding to different warping types and warping levels provided by the embodiment of the present invention;

[0042] Figure 4 is Figure 3 a schematic flow chart of the process of calculating the warping convexity of each sample in each sub-image library in the method. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0044] It should be noted that all directional indications (such as up, down, left, right, front, back...) in the embodiments of the present invention are only used to explain the relative positional relationship and movement conditions between components in a specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indications will also change accordingly.

[0045] It should also be noted that when an element is referred to as being "fixed to" or "disposed on" another element, it can be directly on the other element or there may be an intermediate element at the same time. When an element is referred to as being "connected" to another element, it can be directly connected to the other element or there may be an intermediate element at the same time.

[0046] In addition, the descriptions involving "first", "second", etc. in the present invention are only for descriptive purposes, and cannot be understood as indicating or implying their relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of such features. In addition, the technical solutions between various embodiments can be combined with each other, but it must be based on the fact that those of ordinary skill in the art can implement them. When the combination of technical solutions results in contradictions or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0047] As Figures 1 to 4 shown, it is an on-line recognition method for the warping degree of the cardboard at the outlet of the cardboard production line of an embodiment of the present invention, and is particularly applicable to the cardboard at the outlet of the corrugated cardboard production line.

[0048] Please refer to Figures 1-4 , the on-line recognition method for the warping degree of the cardboard at the outlet of the cardboard production line in the embodiment of the present invention includes the following steps:

[0049] Step S10: Collect the cross-sectional image of the cardboard;

[0050] Step S20: Calculate the warping convexity of the cardboard according to the cross-sectional image;

[0051] Step S30: Determine the quality of the cardboard based on the warping convexity. If the absolute value of the warping convexity is not greater than the preset threshold, it is determined that the quality of the cardboard is qualified. If the absolute value of the warping convexity is greater than the preset threshold, the type and level of the cardboard warping are determined and output according to the value of the warping convexity and the preset range values of the warping convexity corresponding to different warping types and levels.

[0052] Understandably, when judging the quality of the cardboard, first judge whether the quality of the cardboard is qualified or warping occurs. If the quality of the cardboard is qualified, it is determined that the quality is qualified. If the cardboard warps, enter the next step, that is, determine the type and level of the cardboard warping. There are two ways to determine the type and level of the cardboard warping. The first way is to first determine whether the cardboard belongs to the upward warping type or the downward warping type according to the positive or negative of the warping convexity, and then determine which level of the upward warping type or which level of the downward warping type the cardboard belongs to. When determining the level, for the upward warping type, directly compare with the preset range values of the warping convexity of different upward warping levels; for the downward warping type, directly compare with the preset range values of the warping convexity of different downward warping levels. The second way is to directly determine which preset range value of the warping convexity the value of the warping convexity is in, and then look at the corresponding warping type and level, that is, directly compare with all the preset range values of the warping convexity.

[0053] Understandably, the levels of different warping types are set according to production requirements. Classifying different warping types into levels and outputting the warping type and level of the cardboard during on-line identification is to adjust the relevant parameters of the cardboard production line according to different levels in subsequent production control, improve the adjustment accuracy, so that the cardboard produced can meet the quality standard with one or as few adjustments to the cardboard production line as possible, and avoid repeatedly adjusting the cardboard production line many times, thereby improving production efficiency.

[0054] Optionally, the warping types include upward warping and downward warping. The levels of upward warping include at least three types, and the levels of downward warping include at least three types.

[0055] In some embodiments, the upward warping is divided into three levels, namely upward warping - type 1, upward warping - type 2, and upward warping - type 3. Define the warping convexity range value as Then the type and level of the cardboard warping are upward warping - type 1, and the warping convexity range value is Then the type and level of the cardboard warping are upward warping - type 2, and the warping convexity range value is Then the type and level of the cardboard warping are upward warping - type 3.

[0056] Similarly, the downward warping is divided into three levels, namely downward warping - type 1, downward warping - type 2, and downward warping - type 3. Define the warping convexity range value as Then the type and level of the cardboard warping are downward warping - type 1, and the warping convexity range value is defined as The type and level of the cardboard warping are downward warping - type 2, and the defined warping convexity range value is The type and level of the cardboard warping are downward warping - type 3.

[0057] Exemplarily, when the warping convexity is within the range, the output type and level of the cardboard warping are downward warping - type 1.

[0058] The on - line recognition method for the warping degree of the cardboard at the outlet of the cardboard production line in the embodiment of the present invention uses machine vision technology with the characteristics of non - contact, low cost and high precision to collect the cross - sectional image of the outlet cardboard for calculation to obtain the warping convexity of the cardboard, and combines the preset threshold and the preset range values of the warping convexity corresponding to different warping types and warping levels to realize the on - line recognition of the warping degree of the cardboard at the outlet of the cardboard production line. That is, the on - line recognition method for the warping degree of the cardboard at the outlet of the cardboard production line in the embodiment of the present invention realizes the on - line recognition of the warping degree of the cardboard through the means of quantifying the warping degree, solves the problem that the existing cardboard production line depends on manual recognition of the warping degree at the outlet of the cardboard, lays a foundation for realizing the process control of the cardboard quality at the export end of the cardboard, can effectively improve the production efficiency and improve the stability of the cardboard production line.

[0059] It should be noted that in step S10, the cross - sectional image of the cardboard is collected by a camera, and the camera is arranged at the outlet end of the cardboard production line. The focal length can be determined by calculating the straight - line distance according to the position of the measured cardboard and the position of the camera arrangement (the distance between the position of the camera arrangement and the end point of the cardboard ramp is used as the focal length), and the change range of the cardboard order width of the cardboard production line is obtained. Then, the camera is selected for installation according to the above parameters and the cross - sectional image of the cardboard at the outlet end of the cardboard production line is collected. Selecting the camera according to the above parameters can ensure that the collected cross - sectional image has sufficient image clarity and a vertical angle.

[0060] It can be understood that the cross - sectional image of the cardboard can also be collected by other means.

[0061] Please refer to Figure 2 , in step S20, the calculation of the warping convexity of the cardboard includes:

[0062] Step S22: Extract the cross - sectional contour of the cardboard;

[0063] Optionally, the Canny edge detection algorithm is used to extract the cross - sectional contour of the cardboard. It can be understood that other edge detection algorithms can also be used to extract the cross - sectional contour of the cardboard.

[0064] Step S24: Perform data fitting according to the extracted cross - sectional contour to obtain the quadratic function relationship of the cross - sectional contour transverse pixel coordinate - cross - sectional contour longitudinal pixel coordinate. The quadratic function relationship is:

[0065] f(x) = ax 2 + bx + c (1)

[0066] Where x represents the horizontal pixel coordinate of the cross-sectional profile, f(x) represents the vertical pixel coordinate of the cross-sectional profile corresponding to the position of x, and a, b, and c represent constants.

[0067] Exemplarily, the acquisition of the quadratic function relationship is implemented based on the least squares method.

[0068] Step S25: Calculate the warping convexity of the cardboard based on the quadratic function relationship.

[0069] Specifically, the warping convexity of the cardboard is calculated using formula (2), and formula (2) is as follows:

[0070]

[0071] Where δ represents the warping convexity of the cardboard, x a and x b respectively represent the pixel coordinates of the two endpoints of the horizontal pixel coordinate of the quadratic function, and x 1 , x 2 , x 3 respectively represent the pixel coordinates of the three quartering points of the horizontal pixel coordinate of the quadratic function.

[0072] That is, when calculating, the horizontal pixel range of the quadratic function is quartered, and the convexity values at their respective positions are obtained by subtracting the average value of the two endpoints from the middle three quartering points respectively, and then the average value of the above three convexity values is calculated to obtain the warping convexity of the cardboard.

[0073] In the embodiment of the present invention, the warping convexity of the cardboard is calculated through this embodiment, which can not only ensure the accuracy of the calculation result but also simplify the processing steps.

[0074] Please refer to Figure 2 , in some embodiments, the calculation of the warping convexity of the cardboard further includes performing step S21 before extracting the cross-sectional profile of the cardboard: performing image processing on the cross-sectional image, and the image processing includes performing grayscale processing on the sample and extracting the region of interest. Extracting the region of interest is beneficial for the next step of processing, so that the processing time can be reduced and the accuracy can be increased when extracting the cross-sectional profile of the cardboard.

[0075] Please refer to Figure 2, in some embodiments, the calculation of the warping convexity of the cardboard further includes performing step S23 before data fitting based on the extracted cross-sectional profile: performing an image closing operation on the extracted cross-sectional profile, which can bridge narrow discontinuities and small gullies, eliminate small holes, and fill cracks in the contour line to make the cross-sectional profile smooth.

[0076] Please refer to Figure 3 , in some embodiments, the method for obtaining the preset range values of the warping convexity corresponding to different warping types and warping levels is as follows:

[0077] Step S100: Collect cross-sectional images of cardboard with different warping directions and warping degrees as samples;

[0078] Step S200: Identify the warping direction and warping degree of the samples and classify the samples according to different warping types and warping levels to obtain a classification result;

[0079] Step S300: Establish sub-image libraries corresponding to different warping types and warping levels according to the classification results;

[0080] Step S400: Calculate the warping convexity of each sample in each sub-image library one by one, and determine the warping convexity range values corresponding to the warping types and warping levels of each sub-image library according to the minimum and maximum values of the warping convexity of the samples in each sub-image library.

[0081] In the embodiments of the present invention, machine vision technology with the characteristics of non-contact, low cost, and high accuracy is used to collect cross-sectional images of cardboard with different warping directions and warping degrees as samples, and combined with the empirical knowledge of on-site engineers, sub-image libraries corresponding to different warping types and warping levels are constructed. Then, the preset range values of the warping convexity corresponding to different warping types and warping levels are calculated and determined, providing a quantitative basis for the subsequent online identification of the warping degree of the cardboard. Moreover, the method for determining the preset range values of the warping convexity corresponding to different warping types and warping levels can be applied to different cardboard production lines, and the preset range values of the warping convexity corresponding to different warping types and warping levels can be updated according to different production situations.

[0082] In step S10, a cross-sectional image of the cardboard is collected through a camera, and the camera is arranged at the outlet end of the cardboard production line. The focal length can be determined by calculating the straight-line distance based on the position of the measured cardboard and the position of the camera arrangement (the distance between the position of the camera arrangement and the end of the cardboard ramp is used as the focal length), and the change range of the cardboard order width of the cardboard production line is obtained. Then, a camera is selected for installation according to the above parameters and a cross-sectional image of the cardboard at the outlet end of the cardboard production line is collected. Selecting a camera according to the above parameters can ensure that the collected cross-sectional image has sufficient image clarity and a vertical angle.

[0083] In step S200, exemplarily, the on-site engineer identifies the warping direction and degree of the sample based on knowledge and experience, and classifies the samples according to different warping types and levels.

[0084] Please refer to Figure 4 , in some embodiments, the calculation of the warping convexity of each sample includes:

[0085] Step S402: Extract the cross-sectional profile of the cardboard;

[0086] Optionally, the Canny edge detection algorithm is used to extract the cross-sectional profile of the cardboard. It can be understood that other edge detection algorithms can also be used to extract the cross-sectional profile of the cardboard.

[0087] Step S404: Perform data fitting based on the extracted cross-sectional profile to obtain a quadratic function relationship of the cross-sectional profile's horizontal pixel coordinate - cross-sectional profile's vertical pixel coordinate. The quadratic function relationship is:

[0088] f(x') = a'x' 2 + b'x' + c' (3)

[0089] where x' represents the horizontal pixel coordinate of the cross-sectional profile, f(x') represents the vertical pixel coordinate of the cross-sectional profile corresponding to the position of x', and a', b', c' represent constants.

[0090] Exemplarily, the acquisition of the quadratic function relationship is implemented based on the least squares method.

[0091] Step S405: Calculate the warping convexity of the sample based on the quadratic function relationship.

[0092] Specifically, the formula (4) is used to calculate the warping convexity of the sample. The formula (4) is as follows:

[0093]

[0094] where δ' represents the warping convexity of the sample, x' a and x' b respectively represent the pixel coordinates of the two endpoints of the horizontal pixel coordinate of the quadratic function, and x' 1 , x' 2 , x' 3 respectively represent the pixel coordinates of the three quartering points of the horizontal pixel coordinate of the quadratic function.

[0095] That is, when calculating, the horizontal pixel range of the quadratic function is quartered, and the convexity values at their respective positions are obtained by subtracting the average value of the two endpoints from the three middle quartering points respectively, and then the average value of the above three convexity values is calculated to obtain the warping convexity of the sample.

[0096] By calculating the warpage convexity of each sample through this implementation manner, the accuracy of the calculation result can be ensured, and the processing steps can be simplified.

[0097] Exemplarily, when implementing the recognition of the warpage direction and degree of the sample and classifying the sample according to different warpage types and levels to obtain a classification result, the formed classification result includes upward warpage - type 1, upward warpage - type 2, upward warpage - type 3, downward warpage - type 1, downward warpage - type 2, downward warpage - type 3, that is, the upward warpage includes three levels, and the downward warpage also includes three levels. Then, six sub - image libraries are established according to the classification result, and each image library corresponds to upward warpage - type 1, upward warpage - type 2, upward warpage - type 3, downward warpage - type 1, downward warpage - type 2, downward warpage - type 3 respectively. Then, the warpage convexity of each sample in the six sub - image libraries is calculated one by one, and then the warpage convexity range values of upward warpage - type 1, upward warpage - type 2, upward warpage - type 3, downward warpage - type 1, downward warpage - type 2, downward warpage - type 3 are determined, including: determining the warpage convexity range value corresponding to upward warpage - type 1 according to the minimum value and the maximum value of the warpage convexity of the samples in the image library corresponding to upward warpage - type 1 Determining the warpage convexity range value corresponding to upward warpage - type 2 according to the minimum value and the maximum value of the warpage convexity of the samples in the image library corresponding to upward warpage - type 2 Determining the warpage convexity range value corresponding to upward warpage - type 3 according to the minimum value and the maximum value of the warpage convexity of the samples in the image library corresponding to upward warpage - type 3 Determining the warpage convexity range value corresponding to downward warpage - type 1 according to the minimum value and the maximum value of the warpage convexity of the samples in the image library corresponding to downward warpage - type 1 Determining the warpage convexity range value corresponding to downward warpage - type 2 according to the minimum value and the maximum value of the warpage convexity of the samples in the image library corresponding to downward warpage - type 2 Determining the warpage convexity range value corresponding to downward warpage - type 3 according to the minimum value and the maximum value of the warpage convexity of the samples in the image library corresponding to downward warpage - type 3

[0098] Please refer to Figure 4 , in some embodiments, the calculation of the warpage convexity of each sample further includes performing step S401 before extracting the cross - sectional profile of the cardboard: performing image processing on the sample, and the image processing includes performing image grayscale processing and extracting the region of interest on the sample. Extracting the region of interest is beneficial for the next - step processing, so that the processing time can be reduced and the accuracy can be increased when extracting the cross - sectional profile of the cardboard.

[0099] Please refer to Figure 4, in some embodiments, the calculation of the warping convexity of each sample further includes performing step S403 before data fitting according to the extracted cross-sectional profile: performing an image closing operation on the extracted cross-sectional profile, which can bridge narrow discontinuities and small gullies, eliminate small holes, and fill cracks in the contour line to make the cross-sectional profile smooth.

[0100] Understandably, when calculating the warping convexity of each sample, the calculation of one sample can be completed and then the calculation of another sample can be completed, or the first operation can be performed on all samples first, and then the next operation can be performed on all samples until all operations are completed.

[0101] Understandably, the determination of the preset threshold can be determined according to experience. Exemplarily, the preset threshold is 0.01, that is, the cardboard with the absolute value of the warping convexity not greater than 0.01 is qualified in quality. In addition, the determination of the preset threshold can also be obtained by referring to the method for obtaining the preset range value of the warping convexity corresponding to different warping types and warping levels. The preset threshold is obtained by collecting the cross-sectional profile images of the cardboard that meet the quality requirements as samples, calculating the warping convexity of the samples, comparing the absolute values of the calculation results, and taking the maximum absolute value as the preset threshold.

[0102] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structural transformation made by using the content of the specification and drawings of the present invention under the inventive concept of the present invention, or direct / indirect application in other related technical fields is included in the patent protection scope of the present invention.

Claims

1. An on-line recognition method for the warping degree of cardboard at the outlet of a cardboard production line, characterized in that, it includes: Collecting the cross-sectional image of the cardboard; Calculating the warping convexity of the cardboard according to the cross-sectional image; Judging the quality of the cardboard according to the warping convexity. If the absolute value of the warping convexity is not greater than the preset threshold, it is determined that the quality of the cardboard is qualified. If the absolute value of the warping convexity is greater than the preset threshold, the type and level of the cardboard warping are determined and output according to the value of the warping convexity and the preset range values of the warping convexity corresponding to different warping types and warping levels; The calculation of the warping convexity of the cardboard includes: extracting the cross-sectional contour of the cardboard according to the cross-sectional image; performing data fitting on the extracted cross-sectional contour to obtain a quadratic function relationship of the cross-sectional contour horizontal pixel coordinate - cross-sectional contour vertical pixel coordinate, and the quadratic function relationship is: f(x) = ax 2 + bx + c (1) where x represents the horizontal pixel coordinate of the cross-sectional contour, f(x) represents the vertical pixel coordinate of the cross-sectional contour corresponding to the x position, and a, b, and c represent constants; Calculating the warping convexity of the cardboard based on the quadratic function relationship, specifically including: calculating the warping convexity of the cardboard using formula (2), and the formula (2) is as follows: Among them, δ represents the warping convexity of the cardboard, and x a and x b respectively represent the pixel coordinates of the two endpoints of the horizontal pixel coordinates of the quadratic function, and x 1 , x 2 , x 3 respectively represent the pixel coordinates of the three quarter points of the horizontal pixel coordinates of the quadratic function.

2. The on-line recognition method for the warping degree of cardboard at the outlet of a cardboard production line according to claim 1, characterized in that, The cross-sectional image is collected by a camera, and the camera is arranged at the outlet end of the cardboard production line.

3. The on-line recognition method for the warping degree of cardboard at the outlet of a cardboard production line according to claim 1, characterized in that, The calculation of the warping convexity of the cardboard further includes performing image processing on the cross-sectional image before extracting the cross-sectional contour of the cardboard, and the image processing includes performing image grayscale processing and region of interest extraction on the cross-sectional image.

4. The on-line recognition method for the warping degree of cardboard at the outlet of a cardboard production line according to claim 1, characterized in that, The calculation of the warping convexity of the cardboard further includes performing image closing operation processing on the extracted cross-sectional contour before performing data fitting according to the extracted cross-sectional contour.

5. The on-line recognition method for the warping degree of cardboard at the outlet of a cardboard production line according to claim 1, characterized in that, The method for obtaining the preset range values of the warping convexity corresponding to different warping types and warping levels is as follows: Collecting cross-sectional images of cardboard with different warping directions and warping degrees as samples; Identifying the warping direction and warping degree of the samples and classifying the samples according to different warping types and warping levels to obtain classification results; Establishing a sub-image library corresponding to different warping types and warping levels according to the classification results; Calculating the warping convexity of each sample in each sub-image library one by one, and determining the warping convexity range values corresponding to the warping types and warping levels of each sub-image library according to the minimum and maximum values of the warping convexity of the samples in each sub-image library.

6. The on-line recognition method for the warping degree of cardboard at the outlet of a cardboard production line according to claim 5, characterized in that, The calculation of the warping convexity of each sample includes: Extracting the cross-sectional contour of the cardboard; Data fitting is performed based on the extracted cross-sectional profile to obtain a quadratic function relationship between the horizontal pixel coordinates and the vertical pixel coordinates of the cross-sectional profile. The quadratic function relationship is as follows: f(x') = a'x' 2 + b'x' + c' (3) where x' represents the horizontal pixel coordinates of the cross-sectional profile, f(x') represents the vertical pixel coordinates of the cross-sectional profile at the position corresponding to x', and a', b', c' represent constants; The warpage convexity of the sample is calculated based on the quadratic function relationship.

7. The on-line recognition method for the warpage degree of the cardboard at the outlet of the cardboard production line according to claim 6, characterized in that The warpage convexity of the sample is calculated using formula (4), and formula (4) is as follows: Among them, δ' represents the warping convexity of the sample, and x' a and x' b respectively represent the pixel coordinates of the two endpoints of the horizontal pixel coordinates of the quadratic function, and x' 1 , x' 2 , x' 3 respectively represent the pixel coordinates of the three quarter points of the horizontal pixel coordinates of the quadratic function.

8. The on-line recognition method for the warpage degree of the cardboard at the outlet of the cardboard production line according to claim 1, characterized in that The warpage types include upward warping and downward warping. The levels of upward warping include at least three types, and the levels of downward warping include at least three types.

Images